Explanatory Visual Question Answering (EVQA) is a recently proposed multimodal reasoning task consisting of answering the visual question and generating multimodal explanations for the reasoning processes. Unlike traditional Visual Question Answering (VQA) task that only aims at predicting answers for visual questions, EVQA also aims to generate userfriendly explanations to improve the explainability and credibility of reasoning models. To date, existing methods for VQA and EVQA ignore the prompt in the question and enforce the model to predict the probabilities of all answers. Moreover, existing EVQA methods ignore the complex relationships among question words, visual regions, and explanation tokens. Therefore, in this work, we propose a Logic Integrated Neural Inference Network (LININ) to restrict the range of candidate answers based on first-order-logic (FOL) and capture cross-modal relationships to generate rational explanations. Firstly, we design a FOLbased question analysis program to fetch a small number of candidate answers. Secondly, we utilize a multimodal transformer encoder to extract visual and question features, and conduct the prediction on candidate answers. Finally, we design a multimodal explanation transformer to construct cross-modal relationships and generate rational explanations. Comprehensive experiments on benchmark datasets demonstrate the superiority of LININ compared with the state-of-the-art methods for EVQA.
A key object in eXplainable Artificial Intelligence (XAI) is to create intelligent systems capable of reasoning and explaining real-world data to facilitate reliable decision-making. Recent studies have acknowledged the importance of providing user-friendly and verifiable explanations to facilitate trustworthy Visual Question Answering (VQA) systems. This paper aims to promote explainable VQA from both data and method perspectives. First, we propose a new Standard Multimodal Explanation (SME) dataset and a new Few-Shot Multimodal Explanation for VQA (FS-MEVQA) task, which aims to generate the multimodal explanation of the underlying reasoning process for solving visual questions with few training samples. Our SME dataset includes 1,028,230 samples composed of questions, images, answers, and multimodal explanations, which can facilitate research in both traditional MEVQA and FS-MEVQA. To the best of our knowledge, this is the first large-scale dataset with joint language-vision explanations based on standard English and additional visual grounding tokens. Second, we propose a training-free Multimodal Explaining Agent (MEAgent) method based on an LLM agent with multimodal open-world tools to infer answers and generate multimodal explanations for visual questions. Our MEAgent can learn multimodal explanation from merely N(=16) training samples and leverage open-world abilities to perform FS-MEVQA on test samples. Comprehensive experimental results evaluated by language quality metrics, visual detection metric, and visual attribution metrics on our SME dataset indicate the superiority of our method for FS-MEVQA. Our code and data are available at https://github.com/LivXue/FS-MEVQA.
Large Language Models (LLMs) have achieved impressive results in knowledge-based Visual Question Answering (VQA). However existing methods still have challenges: the inability to use external tools autonomously, and the inability to work in teams. Humans tend to know whether they need to use external tools when they encounter a new question, e.g., they tend to be able to give a direct answer to a familiar question, whereas they tend to use tools such as search engines when they encounter an unfamiliar question. In addition, humans also tend to collaborate and discuss with others to get better answers. Inspired by this, we propose the multi-agent voting framework. We design three LLM-based agents that simulate different levels of staff in a team, and assign the available tools according to the levels. Each agent provides the corresponding answer, and finally all the answers provided by the agents are voted to get the final answer. Experiments on OK-VQA and A-OKVQA show that our approach outperforms other baselines by 2.2 and 1.0, respectively.
Text-to-image generative models excel in creating images from text but struggle with ensuring alignment and consistency between outputs and prompts. This paper introduces TextMatch, a novel framework that leverages multimodal optimization to address image-text discrepancies in text-to-image (T2I) generation and editing. TextMatch employs a scoring strategy powered by large language models (LLMs) and visual question-answering (VQA) models to evaluate semantic consistency between prompts and generated images. By integrating multimodal in-context learning and chain of thought reasoning, our method dynamically refines prompts through iterative optimization. This process ensures that the generated images better capture user intent of, resulting in higher fidelity and relevance. Extensive experiments demonstrate that TextMatch significantly improves text-image consistency across multiple benchmarks, establishing a reliable framework for advancing the capabilities of text-to-image generative models. Our code is available at https://anonymous.4open.science/r/TextMatch-F55C/.
We introduce a new task called Defeasible Visual Entailment (DVE), where the goal is to allow the modification of the entailment relationship between an image premise and a text hypothesis based on an additional update. While this concept is well-established in Natural Language Inference, it remains unexplored in visual entailment. At a high level, DVE enables models to refine their initial interpretations, leading to improved accuracy and reliability in various applications such as detecting misleading information in images, enhancing visual question answering, and refining decision-making processes in autonomous systems. Existing metrics do not adequately capture the change in the entailment relationship brought by updates. To address this, we propose a novel inference-aware evaluator designed to capture changes in entailment strength induced by updates, using pairwise contrastive learning and categorical information learning. Additionally, we introduce a reward-driven update optimization method to further enhance the quality of updates generated by multimodal models. Experimental results demonstrate the effectiveness of our proposed evaluator and optimization method.
Visual Language Models have demonstrated remarkable capabilities across tasks, including visual question answering and image captioning. However, most models rely on text-based instructions, limiting their effectiveness in human-machine interactions. Moreover, the quality of language models depends on reasoning and prompting techniques, such as COT, which remain underexplored when using speech instructions. To address these challenges, we propose SilVar, a novel end-to-end multimodal model that uses speech instructions for reasoning in visual question answering. In addition, we investigate reasoning techniques with levels including conversational, simple, and complex speech instruction. SilVar is built upon CLIP, Whisper, and LLaMA 3.1-8B, enabling intuitive interactions by allowing users to provide verbal or text instructions. To this end, we introduce a dataset designed to challenge models with speech-based reasoning tasks for object localization. This dataset enhances the model ability to process and explain visual scenes from spoken input, moving beyond object recognition to reasoning-based interactions. The experiments show that SilVar achieves SOTA performance on the MMMU and ScienceQA benchmarks despite the challenge of speech-based instructions. We believe SilVar will inspire next-generation multimodal reasoning models, toward expert artificial general intelligence. Our code and dataset are available here.
Recent Large Vision-Language Models (LVLMs) have shown promising reasoning capabilities on text-rich images from charts, tables, and documents. However, the abundant text within such images may increase the model's sensitivity to language. This raises the need to evaluate LVLM performance on cross-lingual text-rich visual inputs, where the language in the image differs from the language of the instructions. To address this, we introduce XT-VQA (Cross-Lingual Text-Rich Visual Question Answering), a benchmark designed to assess how LVLMs handle language inconsistency between image text and questions. XT-VQA integrates five existing text-rich VQA datasets and a newly collected dataset, XPaperQA, covering diverse scenarios that require faithful recognition and comprehension of visual information despite language inconsistency. Our evaluation of prominent LVLMs on XT-VQA reveals a significant drop in performance for cross-lingual scenarios, even for models with multilingual capabilities. A mutual information analysis suggests that this performance gap stems from cross-lingual questions failing to adequately activate relevant visual information. To mitigate this issue, we propose MVCL-MI (Maximization of Vision-Language Cross-Lingual Mutual Information), where a visual-text cross-lingual alignment is built by maximizing mutual information between the model's outputs and visual information. This is achieved by distilling knowledge from monolingual to cross-lingual settings through KL divergence minimization, where monolingual output logits serve as a teacher. Experimental results on the XT-VQA demonstrate that MVCL-MI effectively reduces the visual-text cross-lingual performance disparity while preserving the inherent capabilities of LVLMs, shedding new light on the potential practice for improving LVLMs. Codes are available at: https://github.com/Stardust-y/XTVQA.git
Multimodal large language models (MLLMs) have significantly advanced tasks like caption generation and visual question answering by integrating visual and textual data. However, they sometimes produce misleading or hallucinate content due to discrepancies between their pre-training data and real user prompts. Existing approaches using Direct Preference Optimization (DPO) in vision-language tasks often rely on strong models like GPT-4 or CLIP to determine positive and negative responses. Here, we propose a new framework in generating synthetic data using a reward model as a proxy of human preference for effective multimodal alignment with DPO training. The resulting DPO dataset ranges from 2K to 9K image-text pairs, was evaluated on LLaVA-v1.5-7B, where our approach demonstrated substantial improvements in both the trustworthiness and reasoning capabilities of the base model across multiple hallucination and vision-language benchmark. The experiment results indicate that integrating selected synthetic data, such as from generative and rewards models can effectively reduce reliance on human-annotated data while enhancing MLLMs' alignment capability, offering a scalable solution for safer deployment.
Fundus fluorescein angiography (FFA) is critical for diagnosing retinal vascular diseases, but beginners often struggle with image interpretation. This study develops FFA Sora, a text-to-video model that converts FFA reports into dynamic videos via a Wavelet-Flow Variational Autoencoder (WF-VAE) and a diffusion transformer (DiT). Trained on an anonymized dataset, FFA Sora accurately simulates disease features from the input text, as confirmed by objective metrics: Frechet Video Distance (FVD) = 329.78, Learned Perceptual Image Patch Similarity (LPIPS) = 0.48, and Visual-question-answering Score (VQAScore) = 0.61. Specific evaluations showed acceptable alignment between the generated videos and textual prompts, with BERTScore of 0.35. Additionally, the model demonstrated strong privacy-preserving performance in retrieval evaluations, achieving an average Recall@K of 0.073. Human assessments indicated satisfactory visual quality, with an average score of 1.570(scale: 1 = best, 5 = worst). This model addresses privacy concerns associated with sharing large-scale FFA data and enhances medical education.
Recently, Large Language Models (LLMs) have been used for knowledge-based Visual Question Answering (VQA). Despite the encouraging results of previous studies, prior methods prompt LLMs to predict answers directly, neglecting intermediate thought processes. We argue that prior methods do not sufficiently activate the capacities of LLMs. We propose a framework called PLRH that Prompts LLMs with Rationale Heuristics for knowledge-based VQA. The PLRH prompts LLMs with Chain of Thought (CoT) to generate rationale heuristics, i.e., intermediate thought processes, and then leverages the rationale heuristics to inspire LLMs to predict answers. Experiments show that our approach outperforms the existing baselines by more than 2.2 and 2.1 on OK-VQA and A-OKVQA, respectively.
Multimodal learning, a rapidly evolving field in artificial intelligence, seeks to construct more versatile and robust systems by integrating and analyzing diverse types of data, including text, images, audio, and video. Inspired by the human ability to assimilate information through many senses, this method enables applications such as text-to-video conversion, visual question answering, and image captioning. Recent developments in datasets that support multimodal language models (MLLMs) are highlighted in this overview. Large-scale multimodal datasets are essential because they allow for thorough testing and training of these models. With an emphasis on their contributions to the discipline, the study examines a variety of datasets, including those for training, domain-specific tasks, and real-world applications. It also emphasizes how crucial benchmark datasets are for assessing models' performance in a range of scenarios, scalability, and applicability. Since multimodal learning is always changing, overcoming these obstacles will help AI research and applications reach new heights.
Compositional generalization is crucial for artificial intelligence agents to solve complex vision-language reasoning tasks. Neuro-symbolic approaches have demonstrated promise in capturing compositional structures, but they face critical challenges: (a) reliance on predefined predicates for symbolic representations that limit adaptability, (b) difficulty in extracting predicates from raw data, and (c) using non-differentiable operations for combining primitive concepts. To address these issues, we propose NeSyCoCo, a neuro-symbolic framework that leverages large language models (LLMs) to generate symbolic representations and map them to differentiable neural computations. NeSyCoCo introduces three innovations: (a) augmenting natural language inputs with dependency structures to enhance the alignment with symbolic representations, (b) employing distributed word representations to link diverse, linguistically motivated logical predicates to neural modules, and (c) using the soft composition of normalized predicate scores to align symbolic and differentiable reasoning. Our framework achieves state-of-the-art results on the ReaSCAN and CLEVR-CoGenT compositional generalization benchmarks and demonstrates robust performance with novel concepts in the CLEVR-SYN benchmark.
Recent advancements in large vision language models (VLMs) tailored for autonomous driving (AD) have shown strong scene understanding and reasoning capabilities, making them undeniable candidates for end-to-end driving systems. However, limited work exists on studying the trustworthiness of DriveVLMs -- a critical factor that directly impacts public transportation safety. In this paper, we introduce AutoTrust, a comprehensive trustworthiness benchmark for large vision-language models in autonomous driving (DriveVLMs), considering diverse perspectives -- including trustfulness, safety, robustness, privacy, and fairness. We constructed the largest visual question-answering dataset for investigating trustworthiness issues in driving scenarios, comprising over 10k unique scenes and 18k queries. We evaluated six publicly available VLMs, spanning from generalist to specialist, from open-source to commercial models. Our exhaustive evaluations have unveiled previously undiscovered vulnerabilities of DriveVLMs to trustworthiness threats. Specifically, we found that the general VLMs like LLaVA-v1.6 and GPT-4o-mini surprisingly outperform specialized models fine-tuned for driving in terms of overall trustworthiness. DriveVLMs like DriveLM-Agent are particularly vulnerable to disclosing sensitive information. Additionally, both generalist and specialist VLMs remain susceptible to adversarial attacks and struggle to ensure unbiased decision-making across diverse environments and populations. Our findings call for immediate and decisive action to address the trustworthiness of DriveVLMs -- an issue of critical importance to public safety and the welfare of all citizens relying on autonomous transportation systems. Our benchmark is publicly available at \url{https://github.com/taco-group/AutoTrust}, and the leaderboard is released at \url{https://taco-group.github.io/AutoTrust/}.
Multimodal large language models (MLLMs) combine visual and textual data for tasks such as image captioning and visual question answering. Proper uncertainty calibration is crucial, yet challenging, for reliable use in areas like healthcare and autonomous driving. This paper investigates representative MLLMs, focusing on their calibration across various scenarios, including before and after visual fine-tuning, as well as before and after multimodal training of the base LLMs. We observed miscalibration in their performance, and at the same time, no significant differences in calibration across these scenarios. We also highlight how uncertainty differs between text and images and how their integration affects overall uncertainty. To better understand MLLMs' miscalibration and their ability to self-assess uncertainty, we construct the IDK (I don't know) dataset, which is key to evaluating how they handle unknowns. Our findings reveal that MLLMs tend to give answers rather than admit uncertainty, but this self-assessment improves with proper prompt adjustments. Finally, to calibrate MLLMs and enhance model reliability, we propose techniques such as temperature scaling and iterative prompt optimization. Our results provide insights into improving MLLMs for effective and responsible deployment in multimodal applications. Code and IDK dataset: https://github.com/hfutml/Calibration-MLLM.
Large Vision-Language Models typically require large text and image datasets for effective fine-tuning. However, collecting data from various sites, especially in healthcare, is challenging due to strict privacy regulations. An alternative is to fine-tune these models on end-user devices, such as in medical clinics, without sending data to a server. These local clients typically have limited computing power and small datasets, which are not enough for fully fine-tuning large VLMs on their own. A naive solution to these scenarios is to leverage parameter-efficient fine-tuning (PEFT) strategies and apply federated learning (FL) algorithms to combine the learned adapter weights, thereby respecting the resource limitations and data privacy. However, this approach does not fully leverage the knowledge from multiple adapters trained on diverse data distributions and for diverse tasks. The adapters are adversely impacted by data heterogeneity and task heterogeneity across clients resulting in suboptimal convergence. To this end, we propose a novel framework called FedPIA that improves upon the naive combinations of FL and PEFT by introducing Permutation and Integration of the local Adapters in the server and global Adapters in the clients exploiting Wasserstein barycenters for improved blending of client-specific and client-agnostic knowledge. This layerwise permutation helps to bridge the gap in the parameter space of local and global adapters before integration. We conduct over 2000 client-level experiments utilizing 48 medical image datasets across five different medical vision-language FL task settings encompassing visual question answering as well as image and report-based multi-label disease detection. Our experiments involving diverse client settings, ten different modalities, and two VLM backbones demonstrate that FedPIA consistently outperforms the state-of-the-art PEFT-FL baselines.
Artificial intelligence has advanced in Medical Visual Question Answering (Med-VQA), but prevalent research tends to focus on the accuracy of the answers, often overlooking the reasoning paths and interpretability, which are crucial in clinical settings. Besides, current Med-VQA algorithms, typically reliant on singular models, lack the robustness needed for real-world medical diagnostics which usually require collaborative expert evaluation. To address these shortcomings, this paper presents MedCoT, a novel hierarchical expert verification reasoning chain method designed to enhance interpretability and accuracy in biomedical imaging inquiries. MedCoT is predicated on two principles: The necessity for explicit reasoning paths in Med-VQA and the requirement for multi-expert review to formulate accurate conclusions. The methodology involves an Initial Specialist proposing diagnostic rationales, followed by a Follow-up Specialist who validates these rationales, and finally, a consensus is reached through a vote among a sparse Mixture of Experts within the locally deployed Diagnostic Specialist, which then provides the definitive diagnosis. Experimental evaluations on four standard Med-VQA datasets demonstrate that MedCoT surpasses existing state-of-the-art approaches, providing significant improvements in performance and interpretability.
Compositional generalization is the capability of a model to understand novel compositions composed of seen concepts. There are multiple levels of novel compositions including phrase-phrase level, phrase-word level, and word-word level. Existing methods achieve promising compositional generalization, but the consistency of compositional generalization across multiple levels of novel compositions remains unexplored. The consistency refers to that a model should generalize to a phrase-phrase level novel composition, and phrase-word/word-word level novel compositions that can be derived from it simultaneously. In this paper, we propose a meta-learning based framework, for achieving consistent compositional generalization across multiple levels. The basic idea is to progressively learn compositions from simple to complex for consistency. Specifically, we divide the original training set into multiple validation sets based on compositional complexity, and introduce multiple meta-weight-nets to generate sample weights for samples in different validation sets. To fit the validation sets in order of increasing compositional complexity, we optimize the parameters of each meta-weight-net independently and sequentially in a multilevel optimization manner. We build a GQA-CCG dataset to quantitatively evaluate the consistency. Experimental results on visual question answering and temporal video grounding, demonstrate the effectiveness of the proposed framework. We release GQA-CCG at https://github.com/NeverMoreLCH/CCG.
In an effort to create a more efficient AI system, we introduce a new multi-modality learning framework that leverages a modality-agnostic concept space possessing abstract knowledge and a set of modality-specific projection models tailored to process distinct modality inputs and map them onto the concept space. Decoupled from specific modalities and their associated projection models, the concept space focuses on learning abstract knowledge that is universally applicable across modalities. Subsequently, the knowledge embedded into the concept space streamlines the learning processes of modality-specific projection models. We evaluate our framework on two popular tasks: Image-Text Matching and Visual Question Answering. Our framework achieves performance on par with benchmark models while demonstrating more efficient learning curves.
Video text-based visual question answering (Video TextVQA) is a practical task that aims to answer questions by jointly reasoning textual and visual information in a given video. Inspired by the development of TextVQA in image domain, existing Video TextVQA approaches leverage a language model (e.g. T5) to process text-rich multiple frames and generate answers auto-regressively. Nevertheless, the spatio-temporal relationships among visual entities (including scene text and objects) will be disrupted and models are susceptible to interference from unrelated information, resulting in irrational reasoning and inaccurate answering. To tackle these challenges, we propose the TEA (stands for ``\textbf{T}rack th\textbf{E} \textbf{A}nswer'') method that better extends the generative TextVQA framework from image to video. TEA recovers the spatio-temporal relationships in a complementary way and incorporates OCR-aware clues to enhance the quality of reasoning questions. Extensive experiments on several public Video TextVQA datasets validate the effectiveness and generalization of our framework. TEA outperforms existing TextVQA methods, video-language pretraining methods and video large language models by great margins.
Multimodal Large Language Models (MLLMs) have significantly advanced visual tasks by integrating visual representations into large language models (LLMs). The textual modality, inherited from LLMs, equips MLLMs with abilities like instruction following and in-context learning. In contrast, the visual modality enhances performance in downstream tasks by leveraging rich semantic content, spatial information, and grounding capabilities. These intrinsic modalities work synergistically across various visual tasks. Our research initially reveals a persistent imbalance between these modalities, with text often dominating output generation during visual instruction tuning. This imbalance occurs when using both full fine-tuning and parameter-efficient fine-tuning (PEFT) methods. We then found that re-balancing these modalities can significantly reduce the number of trainable parameters required, inspiring a direction for further optimizing visual instruction tuning. We introduce Modality Linear Representation-Steering (MoReS) to achieve the goal. MoReS effectively re-balances the intrinsic modalities throughout the model, where the key idea is to steer visual representations through linear transformations in the visual subspace across each model layer. To validate our solution, we composed LLaVA Steering, a suite of models integrated with the proposed MoReS method. Evaluation results show that the composed LLaVA Steering models require, on average, 500 times fewer trainable parameters than LoRA needs while still achieving comparable performance across three visual benchmarks and eight visual question-answering tasks. Last, we present the LLaVA Steering Factory, an in-house developed platform that enables researchers to quickly customize various MLLMs with component-based architecture for seamlessly integrating state-of-the-art models, and evaluate their intrinsic modality imbalance.